starknet_os: os resources test - invoke tx constant factor

[dorimedini-starkware]

No description provided.

[reviewable-StarkWare]

This change is 

[dorimedini-starkware]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• starknet_os: os resources test - invoke tx constant factor #14146 : 2 dependent PRs (#14147 , #14415 ) 👈 (View in Graphite)
• starknet_os: os resources test - extract setup_test_builder #14145
• starknet_os: os resources test - add secp syscalls #14144 : 1 other dependent PR (#14255 )
• starknet_os: os resources test - add send_message_to_l1 #14143 : 1 other dependent PR (#14153 )
• starknet_os: os resources test - add replace_class #14142
• starknet_os: os resources test - support virtual builtins, add sha256 #14141
• starknet_os: define sha256 batch resources constants #14140
• starknet_os: os resources test - add keccak #14139
• starknet_os: os resources test - add 3 more syscalls #14138
• starknet_os: os resources test - remove all fee transfer syscalls #14137
• starknet_os: os resources test - add emit event #14136
• starknet_os: os resources test - add meta tx #14135
• starknet_os: os resources test - measure linear factor, add deploy #14134
• starknet_os: os resources test - library call #14129
• starknet_os: initial test, measure call contract resources #14128
• blockifier_test_utils: add option to deploy from zero in test accounts #14258
• starknet_os_flow_tests: helper function to declare explicit classes #14257
• starknet_os: os resources test - create fixed deployable account #14173
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[cursor]

PR Summary

Medium Risk
Updates fee-related versioned constants and golden fixtures; incorrect measurements would skew fee estimation and OS resource accounting for invokes.

Overview
Adds test_execute_txs_inner_resources in starknet_os_flow_tests to measure Starknet OS scaffolding for invoke transactions (OS trace minus blockifier validate/execute) and lock the result into versioned constants via expect_file against blockifier_versioned_constants_0_14_4.json.

InvokeFunction execute_txs_inner constants are raised: n_steps 4348 → 4779, range_check_builtin 90 → 110, poseidon_builtin 12 → 11. The 0.14.3→0.14.4 VC diff regression file records these paths. Test setup now funds the stable measurement contract and adds span_calldata for calldata-factor measurements.

Downstream expectations are refreshed for higher invoke fees/resources: RPC EXPECTED_FEE_ESTIMATE, blockifier recursion/step-limit tests, preconfirmed_block.json, and blob generation fixture count.

^{Reviewed by Cursor Bugbot for commit e5e710c. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Artifacts upload workflows:

• Started at 2026-06-08T10:25:06Z

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 214c8b8. Configure here.}

[cursor]

Wrong scaled calldata deduction

Medium Severity

When deriving the InvokeFunction execute_txs_inner constant, the test subtracts the full scaled calldata_factor.resources from the first measurement. With scaling_factor 2, runtime cost for one extended calldata unit uses calculate_resources(1) (half that bundle per field), so the stored constant is too low and misaligns with resources_for_tx_type.

 

^{Reviewed by Cursor Bugbot for commit 214c8b8. Configure here.}

[dorimedini-starkware]

@Yoni-Starkware correct me if I'm wrong but we have to scale up the calldata_factor because when charging for it, we "scale down" by the same amount. with the constant factor, no scaling is required

[dorimedini-starkware]

@dorimedini-starkware made 1 comment.
Reviewable status: 0 of 3 files reviewed, 1 unresolved discussion (waiting on yoavGrs and Yoni-Starkware).

[dorimedini-starkware]

@Yoni-Starkware correct me if I'm wrong but we have to scale up the calldata_factor because when charging for it, we "scale down" by the same amount. with the constant factor, no scaling is required